AM_F2019_03

Tom: Dave was my house leader when I was a freshman. He came here as an undergraduate in 1964, correct? He was president of his fraternity, and the next year he was a house tutor. Back in those days the dean's office had a tutor in every fraternity, and then they cut the budgets and after a while they had almost no house tutors. Okay, until this guy in the early 2000s drank himself to death in the fridge [frat] in October, and they decided oh we should have a house tutor program. But Dave was the house tutor that first year that he graduated as a graduate student, and the next year he was in charge of all the house tutors. He's a very capable person.

He became a vice president at Honeywell, fortune 50 company, before he's 40 years old. He's been CEO of two large chemical companies. He's retired now, but he spent the last ten years sitting on boards of companies and acting as a consultant. He's very outspoken, not quite as outspoken as me because he's a little bit more tactful most of the time, but he will tell you the truth about things. And that's what we just had, that discussion right? And that discussion I just had with you is what I'd like today to be, but I want you to hear it from someone else, so it's not just me who doesn't believe in all these things. But are you just putting up a list? We got additive manufacturing, these are great expectations, everybody says they're going to change the world right?

Is that true or not? They're going to change part of the world, and what you read in the papers is overselling, okay. And you need to think critically about things, okay. They wanted me to show you, remind the students who weren't here yesterday of some of the principles. Yeah, just a few of the high points. For structural material, the two types of materials — second slide — structural materials used in extremely large values, billions of tonnes a year. Functional materials like semiconductors or optical fibers or whatever, used in much smaller volumes. Functional materials could have values of millions of dollars a pound, silicon, okay. Structural materials, they have to be cheap, okay. And so we went through them, we talked about the cost of the material here, the billion tons per year club: stone, concrete, engineering wood, and steel. Talking structural materials, if you're talking about the cost of the material, we went through.

The value of the structural material, the pounds saved over the life of the vehicle: the shipper, a railroad twenty cents a pound; autos two dollars a pound; aircraft two hundred dollars a pound; to get a pound into orbit twenty thousand dollars a pound. When people talk about additive manufacturing, usually they're coming up with examples of these latter, because additive manufacturing is expensive. You're not going to put it in there to build ships or automobiles, okay. Those are the high points you wanted, if you guys — anything more? No, that's good. Okay, so can we go back to the first slide that you had, which was how do we get to venture capital. All right, the last line.

Dave: Whoo. Okay, so when I graduated from MIT in 1968 — that was my undergraduate graduation, my bachelor's degree — the world was a very different place than it is today. We were at the height of a cold war. The government had been for many years extremely interested in investing in advanced materials, and pursuing a degree in sophisticated material science opened up a lot of opportunities. The world's changed significantly since then. Places like Bell Laboratories, which really had some of the most brilliant minds I think in science in the 1960s, disappeared with the breakup of AT&T back in 1970 [1984].

IBM, once a tremendous leader in science, and particularly a company that was focused on using superconductivity to revolutionize the world of computation — the superconducting computer — IBM is just a shadow of its former self and not so much materials focused anymore. So as time has evolved, the question becomes where does the funding come from, who supports new science. And since about 2000 it's been largely venture capital money. People who have a lot of money, don't know what to do with it, and invest. Now, do they invest wisely or not so wisely? One of the VC firms I've worked with in the past pursued what they call the spray-and-pray approach to venture capital. Spread it all out there, throw it over here, throw it over there, and pray for the best result. Serendipitously sometimes that works, most often it doesn't work.

So we see today an interest in private money supporting research because someone is looking for the magic stone, the elixir that's going to create a world-beating new technology in material science. The problem with that, as Tom points out, is anything that has the possibility of being different and can be promoted as being different from some thread of scientific principle becomes a cause célèbre and a focus. And Tom's listed a couple of things up here that I think he's spoken to you about in the past about great expectations. One of the ones that I've been involved with quite recently at a fairly high level is graphene, which I guess is up here. It's the current manifestation of the fact that people who don't know anything have finally discovered that the carbon-carbon bond is really as strong as the carbon-carbon bond is. And there's still a significant amount of interest in pursuing graphene opportunities. Rice University and University of Texas at Austin are probably the two largest focus for technical efforts in the field today in the US, and in Britain it's Manchester University. And that's where you go if you want to see who's got the most current view of graphene.

Problem with graphene: it's a solution looking for a problem. You can speculate, anyone can speculate on what it might be able to do. Number one, it's difficult to produce. Number two, it's difficult to stabilize, because these little carbon sheets that are atomic size barely — they have a lot of surface energy and they don't want to be just little specks of carbon, two-dimensional carbon, floating in some medium. So people have to stabilize them somehow. And the challenge becomes how do you stabilize them, and how do you stabilize them in a way in which they're amenable of processing. Because it's okay to have something that's an intellectual curiosity and submicron in size, carbon sheets, but how do I actually fabricate it, what do I actually do with it? I can think of things I could do with it if I could only somehow process it. So I'm not sure what experience you had, Tom, with the graphene. You stayed away from it? So stories, Tom likes to tell stories.

I've worked on a lot of technologies in my years, some of them sublime and some of them ridiculous. One I did work on was amorphous metals. Amorphous metals were a very big cause célèbre in the 1980-1990 range. It's an interesting concept. As I'm sure most of you know, metals form crystalline structures when they solidify, and the driving force for those crystal structures is thermodynamically very large. The only way that you can create a metallic structure that doesn't have long-range order, that isn't crystalline, is to freeze it very quickly. So in the 70s, in the early 70s here, Nick Grant and others were looking at splat cooling of metals. And the way they did that was to levitate a drop of metal — so you need a magnetic field to levitate it, so we focus basically on iron as your primary component — and then blasted with a jet of air or a jet of nitrogen and smash it up against a really cold surface. Splat cooling was what the people called it. And with that you could make structures that were amorphous. They were metals but they didn't have long-range order, they weren't crystalline.

And people started playing around looking at some of the interesting properties that these materials might have. Probably the most interesting property of these amorphous metals is the fact they have no grain boundaries. And the lack of grain boundaries is very important when you think about things like putting a material in a magnet and subjecting the magnet to alternating field, in other words like a transformer. So if I have a transformer, and it's an alternating current transformer, and I need to reverse the field 60 times a second, I'm gonna discover that the magnetic lines of flux experience hysteresis when they're dragged through the grain boundary. The grain boundary holds them up. But if I eliminate the grain boundary I eliminate the hysteresis loss. So people theorized that they could make very highly efficient transformers using amorphous metals. The problem is, you couldn't make amorphous metal transformers, you couldn't make cores or anything of any substance from a splat-cooled metal. All you had was a metal splat. You can think of it as a pigeon dropping. At least that's the way I used to think about it.

So certain advances in science had to be made in order to turn this — what was then an intellectual curiosity — into something that had the potential for commercial success. The first was, you needed a composition of matter that was more tractable and easier to freeze as an amorphous metal than was currently then known. And the second challenge you had to overcome is, you needed a way to make it economically. And I think Tom's earlier slide here about the cost per pound of making materials is a very important thought you have to carry around whenever you're thinking about a new material in an application. The economics have to be there to make it work.

So someone had to figure out how to create amorphous metals in a form that was low cost. And the fellow who did this was a really bright guy. He basically said, if I could take the splat metal, that little nodule that was floating in space magnetically levitated, I could somehow stabilize it, and if I could somehow pull it away fast enough by winding it up, I could make amorphous metal in a sheet or a ribbon. And what he did was to basically propose the concept of using two surfaces very close to each other, one very cold or refrigerated, another with a nozzle of a small slit, and making what amounted to a sessile drop supported by these two surfaces on a wheel, and ripped this stuff off and wind it up. And long story short, ten years and 100 million dollars later, people figured out how to do that.

The problem is that steel used for transformer applications is cheap. If you look at Tom's slide, steel is one of the cheapest materials up there. So even though there was a new material that had lower hysteresis losses, was it economical enough to pass the test of being a substitute for what was then silicon iron — that's what was used in the cores of transformers — and the answer is, well, marginally. And today the business is about two hundred fifty million dollars in total revenue. It's almost all focused on making transformers for countries where the cost of energy is very high. So the Japanese eventually bought up the technology from the people who had the capability around the world, and they're making the material and making it be successful.

But I think from my own experience working through the issues from a technical perspective of how to create the glassy structure and do it in a way that was at least economically competitive or potentially competitive, in the realm of understanding what the overall economics for the use of the material was — it was a very important thought process. So when Tom talks about his list of Great Expectations, one of the things that people often lose sight of in their enthusiasm and their rush to anoint a new technology as the savior for all of our problems is understanding what the overall economics are to make it, to fabricate it, and then to actually use it and achieve whatever benefit is proposed to be associated with it.

Tom: Okay, he was in charge of Honeywell [Allied Signal] for building Conway, South Carolina, and the Japanese built the plant. And he could tell you stories about all the intellectual property problems of doing Japanese when you're using intellectual property. He can tell you the story about how Allied Signal sent a sample of their little laboratory material before they start building a tonnage plant to a guy named Professor Chad Graham, University of Pennsylvania. You won't tell them that story? You tell them. Okay, well Professor Graham was an expert at measuring magnetic properties. Allied Signal in their research lab didn't have the equipment to measure the magnetic susceptibility of samples, so — they were just down the street in New Jersey from Philadelphia — so they asked Professor Graham, would you guys measure this. He said sure.

So he measured it, found out it had ten times the hysteresis — ten times less hysteresis loss than the silicon iron that Allegheny Ludlum Steel is producing. By the way, Allegheny Ludlum Steel at the time was essentially owned by a guy named Dick Simmons. You've heard of Simmons Hall, okay. Dick Simmons was a graduate of this department, barely got through here with his bachelor's degree, I won't tell you that, but he did graduate. So the problem with Allied Signal — they hadn't signed a nondisclosure agreement with Chad Graham, and so it turns out the University of Pennsylvania and Chad Graham had the intellectual property rights to the low magnetic loss properties of the amorphous metal. And that created a huge problem. And because of that faux pas, there's virtually no one in industry today that doesn't make you sign an intellectual property agreement to talk about what they're doing, okay. That was a hundreds of millions of dollar loss to the company that had developed it, because they didn't think about the intellectual property aspects, okay. So is that fair? You have some more into that story? Dave: No. Fortunately it wasn't me who was steering the ship at that point. I got on board later.

Tom: So the amorphous metals have tremendously low losses, ten times better, and if you did the economics at that time they were going to be all over the United States. General Electric is going to start changing all the transformers on the streets — you know you hear them coming and it's over because the steel plates are vibrating. Except — did you know that Allegheny Ludlum and Arco and a few other steel companies that made the silicon iron — and that was a billion-dollar business — they decided they didn't want to lose it, so they went back to the laboratory and they started improving their product. And all of a sudden the amorphous metals, which didn't have a whole lot of room to get better because they had no grain boundaries — it turns out the others caught up with them, to the point where it's just okay. But if you did the economics when they first discovered the magnetic properties, you would have said silicon iron is dead.

But that's an important lesson I've seen over and over: you can't assume that the competitor is going to remain static. People who have billion-dollar businesses are not willing to — they might not do anything for them when they're fat dumb and happy, but when they see the competition coming they start to think okay. For example, I'd like to give these — the fact that the average productivity gain across all of US manufacturing in the 1980s was 1% growth. In the service sector it was actually negative, because we started using PCs and they actually lowered the productivity. Those old PCs were just a time waste to use, okay. They got a little bit better and so they kind of gotten over that hump, but initially they produced the perfect enemy.

There was one industry that had a 6% per year productivity gain in the 1980s. What industry it was? It was the steel industry, because the American steel industry was being decimated. Doesn't mean they were losing money, but they were gonna die unless they innovated. And it turns out they went from — they had constant production, roughly 100 million tons a year that decade — they went from half a million employees to 250,000 employees. Guess what that is over 10 years? That's a double your productivity, okay. They didn't stay static. And it turns out the mini mills over that period are now the most productive steel companies in the world. The American steel industry is the most productive. They still trade at a discount because Wall Street still thinks they're dogs, and to a certain extent they are dogs, except don't say that to the Tatas — you might know them, the Tatas, right? He's an Indian who was smart enough to know that steel is a big business and all these other people want to get rid of it, he'll take it at ten cents on the dollar, or a penny on the dollar, okay, and now he's a very wealthy man and he owns one of the largest in the world.

Okay, so there are some people who actually understand the technology. And by the way, that guy's Dick Simmons, for which Simmons Hall is named. The American steel industry is losing their shirts in the 1980s. Allegheny Ludlum, run by Dick Simmons, never had an unprofitable quarter. Why? Their CEO understood the business, not just from a business point of view, he knew how to make steel. And I went into the plant — they've had some failures, I'm not gonna play — in general the steel workers and the management hate each other, but you go to Allegheny Ludlum, they hate most of the management, love a Simmons, okay. He's a hero, he saved their jobs and they know it, okay. Questions? You want to talk about additive manufacturing today, you want to picking on that, you want to talk about VCs? Why are the VCs around here? You told you a little bit but we haven't told you the real story.

Dave: So I've been working with VCs and private equity groups quite a bit over the last 10 years. It's kind of been a very interesting experience. Normally I wind up in the conversation when there's something wrong. No one asked me at the beginning how to do it right, they just ask me when something's wrong, can I come and fix it.

And I think today the private investment in commercializing technology is going to continue. I think that the blush is off the rose for materials. I don't think we're going to see a continued focus of investment in the materials area, because it's just becoming more and more difficult to express and create the value equations that we've talked about earlier that satisfy people that they're going to get their 20% return on their investment in three to five years. Any technology that requires physical assets to produce — like mills, big chemical complexes — the cycle lifetime in order to get from start to finish isn't three to five years, it's more like 10 years. And the patience of VC investors is not long enough for that. So we're seeing less and less investment in the material space by VCs today. Basically what we're seeing is a lot of investment by private equity groups who want to buy up assets that are not being used effectively and consolidate them and squeeze out savings by just being more efficient in the way the businesses run.

So one more story about materials. Most of my time right now is spent with carbon fiber reinforced plastics, which is kind of a really interesting snapshot into the value-added side of Tom's slide. 50% of the weight of a Dreamliner 787 is some form of composite material. It's kind of interesting to realize that you're flying on a plastic plane. Most people think about plastics as being not very durable, not highly structural, but when you fly on an A350 — an Airbus A350 or a Boeing 787 — you're flying on a vessel that is 50% by weight composite materials. And composite materials have some attributes. They're lightweight, their strength-to-weight ratio is very good. It happens to be unidirectional, unless you do something to obviate the unidirectionality, because it's made by fibers in a matrix. And the way I have to get around that and make it more isotropic is to lay my fabric, my fibers, up in different patterns so that I get more isotropic behavior. The plastic matrix that's used is an epoxy. Well, it's a thermoset, so when you're processing materials that are thermosets, you better not make a mistake, because what you wind up with — there's a lot of very expensive scrap. You can't reprocess a thermoset.

The challenges that the industry faced today in aerospace, commercial aerospace, are largely overcome, and there probably won't be that much new growth in this industry because the penetration rates are achieving a plateau, unless something fundamentally changes. Now what would drive fundamental change? What would I — if I have a composite material today and its cost is X, what would I think about doing in order to lower the cost of making that material? Anybody got any ideas?

Well, the first thing you need to look at is raw material costs. Today most of the carbon fiber that's used in high-performance structures comes from acrylonitrile, and acrylonitrile is a polymer. It's not a very effective polymer for the production of fiber, because when I create the fiber I actually have to burn off most of the chlorine [nitrogen] and other things that are part of the polymer but not necessarily integral to creating that carbon structure in the fiber. And this is back to Tom's story about the strongest materials, a carbon-carbon bond. Carbon fiber is extremely strong, but in order to get there you have to take a precursor and burn away half its raw material content. So that would be one thing that you need to do to get the cost down.

Another thing you need to do is speed the process up. Fiber manufacturing is an interesting science. I spent most of my professional career as the president of the fibers division of Allied Signal, and making sure that you could run those fiber lines as fast and efficiently as possible was always a challenge. But you could speed the process up, if you were able to do that, and you could significantly reduce the cost of carbon fiber. Today you could then go after the higher volume applications. Looking at Tom's, moving up the volume side, get up to automotive. And so probably today most of the interest in carbon fiber is around finding a recipe for making it cheap enough so that it can go into automotive applications. And what I find interesting is that the leading companies in the world who are looking at automotive applications are German car companies. So something about the value in Germany of a pound saved in a vehicle, particularly an electric vehicle or a hybrid vehicle, is a little bit better than it is other places in the world.

Tom: In fact, that was one of the students in this class — his paper — [the BMW i3] which is a carbon fiber shell. Now that vehicle costs about $40,000, okay. The size of the vehicle, it's a small car, but it turns out the way they were able to do that — plus carbon fibers are kind of expensive these days, then he still got this two-dollar-a-pound savings — is, you actually have knock-on savings. When you went to the carbon fiber you got lighter, which means you could have a smaller battery, which makes you lighter. The knock-on from things. And not only that, he had a manufacturing course with — he took my instruction with Kerry of course, I talked about the value of a pound of weight saved on a turbine disk is not $200 a pound like it is for the structural airframe. It's $2,000 a pound on the engine, and on the disk which spins really fast it's more like 20,000 dollars a pound on the commercial aircraft. So if you take a pound off one you can take a lot of weight off something else. And it turns out taking like 20 pounds out of the engine means lighter wings, and for the air force it means two thousand dollars worth of payload, extra bombs, or extra range in terms of fuel economy.

You have to understand this is an old figure from the 1990s, and you can probably double it or triple it, but a 50 degree Fahrenheit increase in operating — and street thermodynamics, you know the efficiency of even genuine thermodynamics, you worried about delta T over T — back in the 90s that was two billion dollars in fuel savings for the airlines. Fuel was a lot cheaper then, okay, it's probably six billion dollars to get a 50 degree temperature increase. And what have they done to do that today?

They actually cool the turbine blades, so the gas running through the engine is at 3,000 degrees Fahrenheit. What's the melting point of the turbine alloy? 2,400 degrees Fahrenheit. If you didn't have the cooling air going through that turbine blade, your whole engine would melt in seconds. Except the cooling air is built by the compressor which comes from the engine, so if your engine isn't running — or if your engine is running, you will have the cool. But it's an amazing technology if you think about it, to have gas going through there at 600 degrees above the melting point and about 800 degrees above the surface temperature of the alloy. It's amazing engineering.

When the physicists came up with the 20 greatest advances of the 20th century, around 2000, they came up with things like clean water — how big an effect is that around the world, right? They came up with productive agriculture, I mean these people. But number 20, we just barely made it, was turbine blades, okay. Those turbine blades, that industry has really come a long way. It turns out, if you go back and look at the history of MIT, there was a professor here in this department named Schenck, and he was working with Nick Grant on high temperature materials. And he left this place and he went to a place called Pratt Whitney, and he was the guy who actually in the 1960s and 70s developed materials technology for these alloys. So you actually can often go back to MIT. It was an MIT — and Nick Grant — that did this first. Spectrally was a guy at Caltech — no, I said Duwez — what was it, Duwez? Anyway. Dave: Yeah. Tom: Questions? Why are the venture capitalists here if they've only got a three year time horizon? What is their goal?

To fund additive manufacturing technology — the answer is no. Their goal is to get to the IPO. And if you got someone like Tommy who's saying, hey, your powder metallurgy technique for additive manufacturing is a bunch of crap, they want to silence me, because they're not going to have a good IPO if the world learns that what they're selling when they have their IPO is a bunch of crap. But they will walk away when they have their IPO with hundreds of millions of dollars in their pocket, and your grandmothers are going to end up with pensions that are worth a lot less.

So if you would have talked about — right now in the news, Ito and Media Lab, right, they were selling influence. They were selling MIT's reputation, okay. That's not right. But you know what the IPOs and venture capitalists are doing every day around here? They're selling MIT's reputation to Wall Street, and to your grandmothers, for a few hundred bucks and some of them — there are some technologies that really work. I was on engineering council when Akamai went public, and a professor in math became a billionaire overnight, okay. They had the technology, it was a real technology. But as Dave says, when you spray and pray, you don't hit those very often.

Okay, and you can talk about Google and Facebook and things like that, people always like to talk about the big hitters, okay. There are 19 failures for every success — and I'm not talking about the big successes — there are 19 failures for every company that's successful, and most of those are barely making it, okay. And money talks around this place, okay. What you say — and that's what you're hearing in the news where they talk about Epstein — does the upper administration know it? I have no idea, cause I'm not involved in that upper administration anymore. But from what I knew from twenty years ago, I don't doubt it, okay. I've seen people sell their souls.

In fact, there's one that's not up here we could talk about: Molten Metal Technologies. Oh yes. So Molten Metal Technologies, back in the early 90s, there was a guy who was a graduate of chemical engineering, and he hired by US Steel research. And when he went to US Steel research, he learned that liquid iron is a high temperature universal solvent, just like water is a universal solvent at room temperature for all kinds of things. You go to high temperatures, iron will dissolve almost anything. And so he had this great idea when he was at US Steel — he was going to take the steel furnace and throw all the environmental crap into the furnace, okay. Where'd you get an idea? Chase it to US Steel, and US Steel knows something about steel — they know how to melt steel — and they looked at his idea, they said no, we don't want to patent it. He says, well why not? It's — you just don't want to do it because — they remembered, just, if you throw crap in your steel pot, you're going to end up with crap steel out, okay, and you won't be able to make steel. And the people at US Steel thought they wanted to stay in the steel business rather than just make big paperweights, okay.

And so he bought the technology — or he got it, right — and he came back to MIT and he got together with a guy named John Preston, who was president of TLO [Technology Licensing Office]. And the two of them started a company, Molten Metal Technology, and this became one of the darlings of Wall Street overnight. The two of them were worth 30 million apiece, okay, hadn't even built a plant. They got some venture capital money, they built an induction melting facility down in Fall River, Massachusetts, and they did, you know, 2,000 pound heats of steel and throw crap in there and it will all burn up. And you know what you could do with the steel afterwards? You can make big paperweights. They would rust over time. They got so brazen that they claimed — they told the Department of Energy they could take radioactive waste and they could make it non-radioactive by throwing it in steel. How many of you believe that? Wall Street Journal believed it, was on the front page, okay.

Don Sadoway and I were just sitting there shaking our heads saying, they don't understand. A lot of your waste has the elements of chlorine in your waste, and another element is sodium. You know what chlorine does in the steel bath? It combines with the carbon in the steel and the oxygen to form — this Co — CO2. You know what that's called? Phosgene. With nobody — chemical weapon in World War One. And you just have to pour it out of the furnace, okay. Guys, how'd you like to work in that plant, okay. You know what happens when you add sodium to the steel, the molten steel bath? The refractory walls that keep the whole furnace from melting — now the sodium — instead of relining the furnace every three or four years, you need to reline it every week, okay. What a great idea.

Now do you know — the guy was an idiot and they were happy when he quit, and he was happy when he left. And Sadoway and I were just sitting there, huh. And so what happens? Al Gore, vice president of the United States, running for president in the 1990s, so he's gonna come do use the technology, right? He invented the internet, or at least he signed the bill, okay. And he came, and he was gonna come to MIT as part of his campaign and go over to Kresge. They're gonna have a technology roundtable. Who's in charge of the technology roundtable putting together? John Preston, head of MIT's Technology Licensing Office. Would you consider that a conflict, when one of the four companies that he puts on there is Molten Metal something, that he owns 30% of? Now I thought it was a conflict. We wrote to the vice president of research at MIT, said we think this is a conflict. He said don't worry about it. Oh, I won't worry about it.

So you don't think money talks in the upper administration? Okay. No one else there even knew that Preston was putting his company forward as part of this job at MIT with the Vice President of the United States. Anyway. But those types of things happen, folks, okay, and they're gonna happen to you. You're gonna graduate, you're gonna be in some of these positions, and you've got to make a decision now. Are you going to go along with the crowd? Are you going to be willing to resign and go get another job?

I've been working two jobs since I was 19 years old. That's 80 to 100 hours a week. Now I don't work that much anymore, about 70 hours a week. I ran afoul of the MIT administration. I had another job, folks. They tried to push me out of here because I knew the dirt, and I wouldn't keep my mouth shut as department head. They were very happy when I decided to step down, and I decided to step down and keep my mouth shut as long as they don't bother — but if they bother me, I know where the dirt is, okay. So we have sort of an agreement, they leave me alone, I leave them alone, okay. But I don't mind telling you the story, and it'll be on YouTube tomorrow, okay. Anybody wants to watch it, but there's not a million people watching this.

Oh no — they got invited, okay, Don Sadoway and Mickey Klein was not associated when I — Don Sadoway when I read it I blew. First thing I looked at the heart, scanned it quickly. Fortunately MIT is no longer associated with Five, some of the faculty in this department sitting on their scientific advisory board and going around shouting this technology — I think you can get rid of the transuranic elements by just throwing them in the steel bath, oh great, okay. But there were people, faculty, paid $1,000 to not come forward and say — as a scientific advisory board member they couldn't come forward and said no, guys I think you're overstepping, okay. But eventually the Justice Department didn't like — his name once where they're saying the guy's dishonest. He had to resign from the Technology Licensing Office at MIT in disgrace, basically just like Mickey [Joi Ito]. This Epstein thing is open at the moment right now, okay, but it is, it's not a pretty story, okay.

Additive manufacturing, which is what we're supposed to be talking about, is a viable technology. But you should know by now that it's only in the aerospace sector. You're not going to start building auto wheels. And when — next week, if you read the — it's not in the 11 rule things from Digital Alloys, but Desktop Metals, there's an article that I will go through where they came up — have I told you about the eyeglass hinges? Two years and two hundred million dollars worth of research, and the best thing they can come up with — if they can make forty-five thousand eyeglass hinges per hour.

We'll talk about the fallacies of the eyeglass hinges. We have a surface roughness like this — oh, we can put three parts together and turn it into one part, the hinge pin — and it talks in the article about how many people have ever had problems with — yeah, talks about, you won't have that — somebody were all be integrated together, yeah. I don't have a surface roughness like this — what do you think the wear properties of that batter are gonna be? Now you understand why I say some of this is crap, okay. It doesn't take very much critical thinking to realize people are just in this for the money. And your grandmothers are gonna pay for it.

Additive manufacturing — stuff like that — it's like, the way they're laying down one layer at a time. If you watch my casting lectures, I should — did a calculation this years ago. How fast — people are big on vapor depositing and stuff. The maximum growth rate by vapor depositing metal is something on the order of a millimeter per minute. You know, through the kinetic theory of gases, how fast can atoms hit the surface, okay. And it's less than — it's got to half a millimeter. You know how fast you can fill a mold and build up something? It's, just like this is — to show that, I was trying to demonstrate in the casting, and I calculated, I had it somewhere between a hundred thousand and a million fold speed advantage. They talk about speed, okay. Casting a metal is a hundred thousand or a million times faster than vapor deposition, okay.

And additive manufacturing is okay — I'm gonna start with three micron powders rather than pouring a big lump into a bowl, okay. No, you can't always go with the casting if you have to do some other things to it, but even so, when you start off with a hundred thousand or a million fold the advantage, people gotta go a long way. And that machine that will make those forty-five thousand eyeglass hinges in an hour, all right, having four hours, it's a million-dollar machine. And the Keurig machine, you make whole — talk about hinges, but the Keurig chain is a hard drive machine. They can process that material in five minutes.

The first time they ever tried to make a disk by additive manufacturing was 1975. Pratt Whitney had the 75 kilowatt laser, sort of a research thing, sort of, listen, the public domain, but he could really buy one. And he was — they getting big contract to make a six inch turbine disk. Anyway. Okay. Anything else? No? I really — any questions in the class? No? Thanks for coming, Dave. Dave: Thank you, thanks for having me.